A test for the difference of means
- 1 March 1986
- journal article
- research article
- Published by Taylor & Francis in Journal of Statistical Computation and Simulation
- Vol. 24 (1) , 33-46
- https://doi.org/10.1080/00949658608810888
Abstract
This paper offers A new test for the difference of means when the ratio of variances is unknown (the Behrens Fisher problem). The proposed test is akin to the familiar z-test, which is sometimes used when the sample sizes are large, with the exception (hat there is a variance correction factor. Evidence is presented which indicates the proposed test significantly outperforms the test.Keywords
This publication has 10 references indexed in Scilit:
- Size and Power of Tests for Equality of Means of Two Normal Populations with Unequal VariancesJournal of the American Statistical Association, 1975
- Probabilities of the Type I Errors of the Welch Tests for the Behrens-Fisher ProblemJournal of the American Statistical Association, 1971
- Practical Solutions of the Behrens-Fisher ProblemJournal of the American Statistical Association, 1970
- On a Comparison of Means of Two Normal SamplesTheory of Probability and Its Applications, 1968
- Small Sample Behaviour of Certain Tests of the Hypothesis of Equal Means Under Variance HeterogeneityBiometrika, 1960
- Experimental DesignsSoil Science, 1950
- AN EXAMINATION AND FURTHER DEVELOPMENT OF A FORMULA ARISING IN THE PROBLEM OF COMPARING TWO MEAN VALUESBiometrika, 1948
- A Note on the Behrens-Fisher ProblemThe Annals of Mathematical Statistics, 1944
- On Solutions of the Behrens-Fisher Problem, Based on the $t$-DistributionThe Annals of Mathematical Statistics, 1943
- THE FIDUCIAL ARGUMENT IN STATISTICAL INFERENCEAnnals of Eugenics, 1935